artmoehn.com — AI Site Grade

Analysis

Excellent. I now have a comprehensive picture. Let me compile the audit.

Cold-Knowledge Geography Error

A frontier LLM queried cold about Art Moehn Auto Group places the dealership group in Greater Milwaukee, Wisconsin — operating Chevrolet, Buick, GMC, and Cadillac brands in Waukesha and nearby communities. The actual site is a Michigan-based Chevrolet and Honda dealer with locations in Jackson and Chelsea, founded in 1962 in Jackson, MI. The model hallucinated an entirely different state, brand mix, and market footprint.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Applebot-Extended, Bytespider, ChatGPT-User, anthropic-ai — receive a 200 with full HTML content on the homepage. No UA-based blocking exists. The robots.txt contains zero AI-bot directives: it blocks MJ12bot, Mail.RU_Bot, AhrefsBot, Baiduspider, Yandex, and a few others, but no GPTBot, ClaudeBot, or any AI crawler is mentioned. No llms.txt exists (404). The site runs on a bare Apache/2.4.41 (Ubuntu) server at DigitalOcean (167.99.187.246) with no CDN or WAF — no Cloudflare, no Akamai, no security headers (no HSTS, no CSP, no X-Frame-Options). The separate brand subdomains (artmoehnchevy.com, artmoehnhonda.com, artmoehnchelsea.com) all return 403 Access Denied via Akamai CDN, blocking even browser UAs — a fragmented infrastructure where the main site is wide open but the brand-specific sites are locked behind a WAF.

Schema and Content Posture

The homepage carries a single AutoDealer JSON-LD block with name, address, phone, and geo coordinates — correct for the Jackson location. Inventory detail pages add richer Car + Offer + Brand schema with VIN, price, mileage, and trim data. However, the about, reviews, blog, and why-buy pages carry zero schema markup — no LocalBusiness, no Review, no Article, no BreadcrumbList. The blog has only three posts (Feb–May 2026), all generic Michigan-driver content. No FAQ schema, no comparison tables, no structured answer-format signals exist anywhere on the site.

External Signals Gap

Web searches for "Art Moehn Auto Group" return zero indexed results from DuckDuckGo — no Google review pages, no Reddit threads, no press mentions, no dealer-review aggregator pages surfaced. The site claims "over 1,000 Google reviews" and "4.5/5 customer rating" on the homepage, but no external citation or review-widget embed is present to substantiate this. The collision center subdomain (artmoehnbodyshop.com) is the only external-facing brand site that loads (200) and includes Google review quotes, Facebook/Instagram/TikTok social links, and a Sunbit financing portal.

Fragmented Domain Architecture

The main site (artmoehn.com) is built on the LiftKit platform (a FlexDealer product) and serves full HTML to all crawlers. But the brand-specific subdomains (artmoehnchevy.com, artmoehnhonda.com, artmoehnchelsea.com) — which are linked from the main site's navigation and inventory pages — are inaccessible to any crawler (403 via Akamai). Inventory VDP pages on the main site reference artmoehnchevy.com in their offers.url schema field, creating a canonical citation pointing to a blocked domain. AI engines crawling the main site will encounter schema that directs them to a URL they cannot access.

Findings

1. Cold-knowledge hallucination places dealership in wrong state and brand mix High

   A frontier LLM queried cold about Art Moehn Auto Group incorrectly placed the dealership in Greater Milwaukee, Wisconsin, with a different brand mix, rather than its actual Michigan-based Chevrolet and Honda operations.

   What to change: Publish a structured knowledge graph (e.g., llms.txt, Wikipedia, or a dedicated about page with schema) to ground AI models in accurate location and brand data.

2. Brand-specific subdomains return 403 to all crawlers High

   The subdomains artmoehnchevy.com, artmoehnhonda.com, and artmoehnchelsea.com return 403 Access Denied via Akamai CDN, blocking all crawlers including AI bots. These subdomains are linked from the main site's navigation and inventory pages.

   What to change: Allow AI crawler access to brand subdomains by adjusting Akamai WAF rules, or consolidate all inventory content under the main domain.

3. Inventory schema references blocked subdomains in offers.url High

   Inventory VDP pages on the main site include schema with offers.url pointing to artmoehnchevy.com, which is blocked to all crawlers. AI engines following these links encounter a 403, breaking the structured data trail.

   What to change: Update offers.url in inventory schema to point to the main domain or ensure the referenced subdomain is accessible to crawlers.

4. Zero external citations found for dealership name High

   Web searches for 'Art Moehn Auto Group' and related queries returned no indexed results on DuckDuckGo, including no Google review pages, press mentions, or dealer-review aggregator pages.

   What to change: Build external citations by claiming and verifying Google Business Profile, Yelp, DealerRater, and other review platforms, and encourage customer reviews.

5. About, reviews, blog, and why-buy pages lack schema markup Medium

   Pages such as about-us/our-history, about-us/our-reviews, blog, and about-us/why-buy contain no structured data (no LocalBusiness, Review, Article, or BreadcrumbList schema), reducing their visibility to AI engines.

   What to change: Add appropriate JSON-LD schema (LocalBusiness, Review, Article, BreadcrumbList) to these pages.

6. No llms.txt file published Medium

   The site returns a 404 for llms.txt, missing an opportunity to provide AI crawlers with a curated summary of key pages and content.

   What to change: Create and publish an llms.txt file listing important pages and a brief site description.

7. No FAQ or comparison schema on the site Medium

   The site lacks any FAQ schema or comparison tables, which are commonly used by AI assistants to answer user questions about dealership services and inventory.

   What to change: Add FAQ schema to relevant pages (e.g., about, why-buy) and consider adding comparison tables for vehicle models.

8. Blog has only three posts with generic content Low

   The blog contains only three posts from Feb–May 2026, all generic Michigan-driver content, providing little unique or authoritative material for AI models to reference.

   What to change: Expand the blog with more frequent, locally relevant content that showcases expertise and community involvement.

9. Main site lacks security headers Low

   The main site (artmoehn.com) runs on bare Apache with no HSTS, CSP, or X-Frame-Options headers, which may affect trust signals for some AI crawlers.

   What to change: Add security headers such as HSTS, CSP, and X-Frame-Options to improve trustworthiness.

What's working

• All major AI crawlers receive full HTML content — The homepage returns 200 with full HTML to GPTBot, ClaudeBot, PerplexityBot, and other AI crawlers, with no UA-based blocking.
• Inventory detail pages include rich Car and Offer schema — Inventory VDP pages contain JSON-LD with Car, Offer, and Brand schema including VIN, price, mileage, and trim data, providing structured information to AI engines.
• Homepage includes correct AutoDealer JSON-LD — The homepage carries a single AutoDealer JSON-LD block with accurate name, address, phone, and geo coordinates for the Jackson location.
• robots.txt does not block any AI crawlers — The robots.txt file contains no directives targeting GPTBot, ClaudeBot, or any AI crawler, ensuring they can access the site freely.
• Collision center subdomain is accessible and includes social links — The artmoehnbodyshop.com subdomain loads successfully (200) and includes Google review quotes, social media links, and financing portal, providing some external signals.